(95) Behavioral Addictions Might Be More Common and Harmful Than Previously Thought

Background & Introduction: The term “addiction” has been inconsistently defined, and there has been ongoing debate in published literature about which problems should be included as addictive disorders in the DSM (e.g., caffeine use disorder, internet gaming disorder, and hypersexual disorder were considered for inclusion in the DSM-5 but were not included). The aims of the present study were: (a) to use a new online addiction questionnaire to assess the current prevalence of 23 possible addictions - 13 substance-related addictions (all in the DSM-5) and 10 behavioral addictions (9 conditions commonly referred to addictions that are not in the DSM-5, plus gambling, which is in the DSM-5) , (b) advise participants about whether they should seek professional help for their addiction-related problem(s), (c) compare the current prevalence of substance and behavioral addictions, and (d) compare self-reported harmful outcomes of substance vs. behavioral addictions.

Methods: We analyzed data obtained from a diverse group of 561 individuals who had completed a new questionnaire online. We evaluated the predictive validity of the questionnaire following guidelines for a "concurrent study design," as described in the most recent version of the Standards for Educational and Psychological Testing, published jointly by the American Psychological Association, the American Educational Research Association, and the National Council on Measurement in Education. For validity evidence, we looked for associations between test scores and answers to 10 different criterion questions. We used Cronbach’s alpha and the Guttman split-half test to estimate the internal consistency of the questionnaire. For demographic comparisons: With two groups, we used a t-test to compare means; with three or more groups, we used an analysis of variance (ANOVA). When comparing two means, we also reported Cohen’s d as an estimate of effect size. When comparing three or more means, we reported eta squared (η²); eta squared was only reported when the p value was significant. We used a Pearson r to measure associations between test scores and various life outcomes (as determined by answers to our criterion questions). We evaluated possible behavioral and substance addictions separately using these and other measures.

Results: Correlations between answers to our 10 criterion questions and scores on behavioral addictions were all significant at the .001 level. In other words, test scores were predictive of a wide variety of life outcomes (no causal relationships are implied here). Test reliability, as measured by internal consistency, was strong (Cronbach’s alpha = 0.87; Guttman split-half = 0.84). As people got older, the prevalence of addictions (as indicated by test scores) decreased (r = -0.28, p < .001). Our two most important findings, we believe, were as follows: (1) The prevalence of possible behavioral addictions (0.085) proved to be significantly greater than the prevalence of possible substance addictions (0.030) (t(560) = 12.54, p < .001, d = 0.53). (2) Associations between harmful life outcomes and possible behavioral addictions (r = -0.37) proved to be substantially and significantly greater than associations between harmful life outcomes and possible substance addictions (r = -0.14, z = 4.10, p < .001).

Conclusion & Discussion: Our data suggest (1) that behavioral addictions might now be far more common than substance addictions, perhaps because of advances in technology in recent decades (creating possible addictions such as internet gaming addiction) and, of greater importance, (2) that behavioral addictions might now be producing more harmful life outcomes than substance addictions do. If these findings hold up to future study, the health and mental health professions might need to reassess the way they currently classify and prioritize a wide variety of possible addictions. The main limitation of our study is its relatively small sample, which we are currently in the process of expanding. If our findings hold up with a much larger sample, the concerns we have raised about the current classifications of addictions will need to be examined more closely. Ultimately, they might have implications for both classifications and treatment priorities.

References: Alimoradi Z, Lin CY, Broström A, et al. Estimation of behavioral addiction prevalence during the COVID-19 pandemic: a systematic review and meta-analysis. Curr Opin Psychiatry. 2022;35(4):317-328. doi:10.1097/YCO.0000000000000826

Kotyuk E, Magi A, Eisinger A, et al. Co-occurrences of substance use and other potentially addictive behaviors: Epidemiological results from the Psychological and Genetic Factors of the Addictive Behaviors (PGA) Study. J Behav Addict. 2020;9(2):272-288. Published 2020 Jun 26. doi:10.1556/2006.2020.00033

Potenza MN. Non-substance addictive behaviors in the context of DSM-5. Addiction. 2013;108(3):480-492. doi:10.1111/add.12117